Re^2: Predictable random sequence

Replies are listed 'Best First'.
Re^3: Predictable random sequence by BrowserUk (Patriarch) on Sep 17, 2013 at 14:49 UTC
but seems it's not cleared in documentation that sequence is always same for same seed. It does, else it wouldn't be the Mersenne Twister. Update: Results from a 32-bit and a 64-bit build of Perl (on windows): C:\test>\perl32\bin\perl -MMath::Random::MT=rand,srand -le"srand(1); p +rint rand(12345) for 1 .. 50" 0.417021998437122 0.99718480813317 0.720324489288032 0.932557361200452 + 0.000114381080493331 0.128124447772279 0.302332567749545 0.999040515394881 0.14675589255057 +3 0.236088976264 0.0923385955393314 0.39658072614111 0.186260211281478 0.38791074021719 +4 0.345560725079849 0.66974604036659 0.396767468657345 0.935539072612301 0.538816731888801 + 0.846310918219388 0.419194519054145 0.313273512991145 0.685219500679523 0.52454816270619 +6 0.204452248988673 0.44345289375633 0.878117437008768 0.22957722004503 0.027387595968321 +0.534413907211274 0.670467507559806 0.913962022401392 0.417304804315791 0.45720480917953 + 0.558689827332273 0.430698570096865 0.140386936953291 0.939127794932574 0.19810148328542 +7 0.778389235492796 0.800744566367939 0.715970510849729 0.968261571833864 0.80275750323198 +7 0.313424184685573 0.0928008102346212 0.692322616931051 0.518152548233047 0.8763891460839 +66 0.865020245313644 C:\test>\perl64\bin\perl -MMath::Random::MT=rand,srand -le"srand(1); p +rint rand(12345) for 1 .. 50" 0.417021998437122 0.99718480813317 0.720324489288032 0.932557361200452 + 0.000114381080493331 0.128124447772279 0.302332567749545 0.999040515394881 0.14675589255057 +3 0.236088976264 0.0923385955393314 0.39658072614111 0.186260211281478 0.38791074021719 +4 0.345560725079849 0.66974604036659 0.396767468657345 0.935539072612301 0.538816731888801 + 0.846310918219388 0.419194519054145 0.313273512991145 0.685219500679523 0.52454816270619 +6 0.204452248988673 0.44345289375633 0.878117437008768 0.22957722004503 0.027387595968321 +0.534413907211274 0.670467507559806 0.913962022401392 0.417304804315791 0.45720480917953 + 0.558689827332273 0.430698570096865 0.140386936953291 0.939127794932574 0.19810148328542 +7 0.778389235492796 0.800744566367939 0.715970510849729 0.968261571833864 0.80275750323198 +7 0.313424184685573 0.0928008102346212 0.692322616931051 0.518152548233047 0.8763891460839 +66 0.865020245313644 [download] You should get the same everywhere regardless of platform or build. With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday' Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice.	[reply] [d/l]
Re^4: Predictable random sequence by vsespb (Chaplain) on Sep 17, 2013 at 15:10 UTC
I am getting different results on linux: $ perl -MMath::Random::MT=rand,srand -le"srand(1); print rand(12345) f +or 1 .. 50" 5148.13657070627 12310.246456404 8892.40582026076 11512.4206240196 1.41203443869017 1581.69630774879 3732.29554886813 12333.1551625498 1811.70149353682 2914.51841197908 1139.91996193305 4895.789064212 2299.38230826985 4788.75808798126 4265.94715111074 8268.01486832555 4898.09440057492 11549.2298513988 6651.69255516725 10447.7082854183 5174.95633772342 3867.36151787569 8459.03473588871 6475.54706860799 2523.96301376517 5474.42597342189 10840.3597598732 2834.1307814559 338.099872228922 6597.33968452318 8276.92138082581 11282.8611665452 5151.62780927843 5644.19336932129 6897.02591841691 5316.9738478458 1733.07673668838 11593.5326284426 2445.5628111586 9609.21511215856 9885.19167181221 8838.65595643991 11953.189104289 9910.04137739888 3869.2215599434 1145.6260023464 8546.72270601382 6396.59320793697 10819.0240084066 10678.6749283969 [download] Maybe you meant this? $ perl -MMath::Random::MT=rand,srand -le"srand(1); print rand() for 1 +.. 50" 0.417021998437122 0.99718480813317 0.720324489288032 0.932557361200452 0.000114381080493331 0.128124447772279 0.302332567749545 0.999040515394881 0.146755892550573 0.236088976264 0.0923385955393314 0.39658072614111 0.186260211281478 0.387910740217194 0.345560725079849 0.66974604036659 0.396767468657345 0.935539072612301 0.538816731888801 0.846310918219388 0.419194519054145 0.313273512991145 0.685219500679523 0.524548162706196 0.204452248988673 0.44345289375633 0.878117437008768 0.22957722004503 0.027387595968321 0.534413907211274 0.670467507559806 0.913962022401392 0.417304804315791 0.45720480917953 0.558689827332273 0.430698570096865 0.140386936953291 0.939127794932574 0.198101483285427 0.778389235492796 0.800744566367939 0.715970510849729 0.968261571833864 0.802757503231987 0.313424184685573 0.0928008102346212 0.692322616931051 0.518152548233047 0.876389146083966 0.865020245313644 [download] Anyway, even if numbers are same, point was it's not documented in module .pod, so might change in the future/in the past. Who knows. UPD: seems second example same as yours.	[reply] [d/l] [select]
Re^5: Predictable random sequence by BrowserUk (Patriarch) on Sep 17, 2013 at 15:32 UTC
Anyway, even if numbers are same, point was it's not documented in module .pod, so might change in the future/in the past. If it does/did, it would not be a faithful implementation of the Mersenne twister. UPD: seems second example same as yours. It appears that the argument to rand is being ignored on both my windows installations but not on your linux install. Which is intriguing as they should be using the same perl code. I don't have an explanation for that yet. But, it you use it without an argument (ie. you retrieve just the 0-1 range random float and multiply it out yourself), you'll get the same results everywhere. That doesn't need to be documented because it is a fact of the design of the MT. With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday' Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice.	[reply]
Re^5: Predictable random sequence by BrowserUk (Patriarch) on Sep 17, 2013 at 15:50 UTC
Hm. Intriguing. It seems that the Windows versions are doing the 'right thing'. Ie. ignoring the argument when you call rand and returning a random float 'tween 0 .. 1. To get the equivalent behaviour of the built-in rand where the random float is multiplied by the argument, you need to import the function irand(). I cannot see by inspection why (or even how) you get different behaviour on linux? It seems it should do the same thing as the Windows install: `sub rand { my ($self, $N) = @_; unless (ref $self) { Math::Random::MT::srand() unless defined $gen; $self = $gen; } return $self->genrand(); } sub irand { my ($self, $N) = @_; unless (ref $self) { $N = $self; Math::Random::MT::srand() unless defined $gen; $self = $gen; } return ($N \|\| 1) * $self->genirand(); }` [download] With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday' Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice.	[reply] [d/l]
Re^6: Predictable random sequence by choroba (Cardinal) on Sep 17, 2013 at 16:06 UTC
Re^7: Predictable random sequence by BrowserUk (Patriarch) on Sep 17, 2013 at 16:31 UTC
Some notes below your chosen depth have not been shown here
Re^4: Predictable random sequence by vsespb (Chaplain) on Sep 18, 2013 at 15:38 UTC
It does, else it wouldn't be the Mersenne Twister. Which one? here are at least two common variants of the algorithm, differing only in the size of the Mersenne primes used. The newer and more commonly used one is the Mersenne Twister MT19937, with 32-bit word length. There is also a variant with 64-bit word length, MT19937-64, which generates a different sequence.	[reply]
Re^5: Predictable random sequence by BrowserUk (Patriarch) on Sep 18, 2013 at 15:59 UTC
Which one? The one used by Math::Random::MT :) (Specifically, the version documented in the source code for the module.) There are actually several variants of the MT, but the module only implements one of them, and it (the module) therefore produces the same sequences everywhere it compiles, regardless of whether it is compiled on a 32-bit or 64-bit platform; or what the underlying OS is. That is its main strength for simulation work. With the rise and rise of 'Social' network sites: 'Computers are making people easier to use everyday' Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error. "Science is about questioning the status quo. Questioning authority". In the absence of evidence, opinion is indistinguishable from prejudice.	[reply]


"be consistent"
	PerlMonks